X-Git-Url: https://git.openssl.org/gitweb/?p=openssl.git;a=blobdiff_plain;f=crypto%2Fsha%2Fsha512.c;h=aab1f1e8d457931fac527beddd9cdc74c6a7ff6c;hp=e48e2febbe3b5a0803002ce50d38ca10b563648a;hb=da3bd2779dffae5980ebfb291c82ad8aba0b8e74;hpb=ce6aa1e49607b21c0cb712f38b97d8fa80f6051d

diff --git a/crypto/sha/sha512.c b/crypto/sha/sha512.c
index e48e2febbe..aab1f1e8d4 100644
--- a/crypto/sha/sha512.c
+++ b/crypto/sha/sha512.c
@@ -50,9 +50,13 @@
 
 #include "cryptlib.h"
 
-const char *SHA512_version="SHA-512" OPENSSL_VERSION_PTEXT;
+__fips_constseg
+const char SHA512_version[]="SHA-512" OPENSSL_VERSION_PTEXT;
 
-#if defined(_M_IX86) || defined(_M_AMD64) || defined(__i386) || defined(__x86_64)
+#if defined(__i386) || defined(__i386__) || defined(_M_IX86) || \
+    defined(__x86_64) || defined(_M_AMD64) || defined(_M_X64) || \
+    defined(__s390__) || defined(__s390x__) || \
+    defined(SHA512_ASM)
 #define SHA512_BLOCK_CAN_MANAGE_UNALIGNED_DATA
 #endif
 
@@ -66,6 +70,7 @@ int SHA384_Init (SHA512_CTX *c)
 	c->h[5]=U64(0x8eb44a8768581511);
 	c->h[6]=U64(0xdb0c2e0d64f98fa7);
 	c->h[7]=U64(0x47b5481dbefa4fa4);
+
         c->Nl=0;        c->Nh=0;
         c->num=0;       c->md_len=SHA384_DIGEST_LENGTH;
         return 1;
@@ -81,6 +86,7 @@ int SHA512_Init (SHA512_CTX *c)
 	c->h[5]=U64(0x9b05688c2b3e6c1f);
 	c->h[6]=U64(0x1f83d9abfb41bd6b);
 	c->h[7]=U64(0x5be0cd19137e2179);
+
         c->Nl=0;        c->Nh=0;
         c->num=0;       c->md_len=SHA512_DIGEST_LENGTH;
         return 1;
@@ -89,7 +95,7 @@ int SHA512_Init (SHA512_CTX *c)
 #ifndef SHA512_ASM
 static
 #endif
-void sha512_block (SHA512_CTX *ctx, const void *in, size_t num);
+void sha512_block_data_order (SHA512_CTX *ctx, const void *in, size_t num);
 
 int SHA512_Final (unsigned char *md, SHA512_CTX *c)
 	{
@@ -100,7 +106,7 @@ int SHA512_Final (unsigned char *md, SHA512_CTX *c)
 	n++;
 	if (n > (sizeof(c->u)-16))
 		memset (p+n,0,sizeof(c->u)-n), n=0,
-		sha512_block (c,p,1);
+		sha512_block_data_order (c,p,1);
 
 	memset (p+n,0,sizeof(c->u)-16-n);
 #ifdef	B_ENDIAN
@@ -125,7 +131,7 @@ int SHA512_Final (unsigned char *md, SHA512_CTX *c)
 	p[sizeof(c->u)-16] = (unsigned char)(c->Nh>>56);
 #endif
 
-	sha512_block (c,p,1);
+	sha512_block_data_order (c,p,1);
 
 	if (md==0) return 0;
 
@@ -191,13 +197,13 @@ int SHA512_Update (SHA512_CTX *c, const void *_data, size_t len)
 
 		if (len < n)
 			{
-			memcpy (p+c->num,data,len), c->num += len;
+			memcpy (p+c->num,data,len), c->num += (unsigned int)len;
 			return 1;
 			}
 		else	{
 			memcpy (p+c->num,data,n), c->num = 0;
 			len-=n, data+=n;
-			sha512_block (c,p,1);
+			sha512_block_data_order (c,p,1);
 			}
 		}
 
@@ -207,12 +213,12 @@ int SHA512_Update (SHA512_CTX *c, const void *_data, size_t len)
 		if ((size_t)data%sizeof(c->u.d[0]) != 0)
 			while (len >= sizeof(c->u))
 				memcpy (p,data,sizeof(c->u)),
-				sha512_block (c,p,1),
+				sha512_block_data_order (c,p,1),
 				len  -= sizeof(c->u),
 				data += sizeof(c->u);
 		else
 #endif
-			sha512_block (c,data,len/sizeof(c->u)),
+			sha512_block_data_order (c,data,len/sizeof(c->u)),
 			data += len,
 			len  %= sizeof(c->u),
 			data -= len;
@@ -227,7 +233,7 @@ int SHA384_Update (SHA512_CTX *c, const void *data, size_t len)
 {   return SHA512_Update (c,data,len);   }
 
 void SHA512_Transform (SHA512_CTX *c, const unsigned char *data)
-{   sha512_block (c,data,1);  }
+{   sha512_block_data_order (c,data,1);  }
 
 unsigned char *SHA384(const unsigned char *d, size_t n, unsigned char *md)
 	{
@@ -256,6 +262,7 @@ unsigned char *SHA512(const unsigned char *d, size_t n, unsigned char *md)
 	}
 
 #ifndef SHA512_ASM
+__fips_constseg
 static const SHA_LONG64 K512[80] = {
         U64(0x428a2f98d728ae22),U64(0x7137449123ef65cd),
         U64(0xb5c0fbcfec4d3b2f),U64(0xe9b5dba58189dbbc),
@@ -301,7 +308,7 @@ static const SHA_LONG64 K512[80] = {
 #ifndef PEDANTIC
 # if defined(__GNUC__) && __GNUC__>=2 && !defined(OPENSSL_NO_ASM) && !defined(OPENSSL_NO_INLINE_ASM)
 #  if defined(__x86_64) || defined(__x86_64__)
-#   define ROTR(a,n)	({ unsigned long ret;		\
+#   define ROTR(a,n)	({ SHA_LONG64 ret;		\
 				asm ("rorq %1,%0"	\
 				: "=r"(ret)		\
 				: "J"(n),"0"(a)		\
@@ -315,29 +322,30 @@ static const SHA_LONG64 K512[80] = {
 #  elif (defined(__i386) || defined(__i386__)) && !defined(B_ENDIAN)
 #   if defined(I386_ONLY)
 #    define PULL64(x) ({ const unsigned int *p=(const unsigned int *)(&(x));\
-			unsigned int hi,lo;			\
+			 unsigned int hi=p[0],lo=p[1];		\
 				asm("xchgb %%ah,%%al;xchgb %%dh,%%dl;"\
 				    "roll $16,%%eax; roll $16,%%edx; "\
 				    "xchgb %%ah,%%al;xchgb %%dh,%%dl;" \
 				: "=a"(lo),"=d"(hi)		\
-				: "0"(p[1]),"1"(p[0]) : "cc");	\
+				: "0"(lo),"1"(hi) : "cc");	\
 				((SHA_LONG64)hi)<<32|lo;	})
 #   else
 #    define PULL64(x) ({ const unsigned int *p=(const unsigned int *)(&(x));\
-			unsigned int hi,lo;			\
+			 unsigned int hi=p[0],lo=p[1];		\
 				asm ("bswapl %0; bswapl %1;"	\
 				: "=r"(lo),"=r"(hi)		\
-				: "0"(p[1]),"1"(p[0]));		\
+				: "0"(lo),"1"(hi));		\
 				((SHA_LONG64)hi)<<32|lo;	})
 #   endif
-#  elif defined(_ARCH_PPC) && defined(__64BIT__)
-#   define ROTR(a,n)	({ unsigned long ret;		\
+#  elif (defined(_ARCH_PPC) && defined(__64BIT__)) || defined(_ARCH_PPC64)
+#   define ROTR(a,n)	({ SHA_LONG64 ret;		\
 				asm ("rotrdi %0,%1,%2"	\
 				: "=r"(ret)		\
 				: "r"(a),"K"(n)); ret;	})
 #  endif
 # elif defined(_MSC_VER)
 #  if defined(_WIN64)	/* applies to both IA-64 and AMD64 */
+#   pragma intrinsic(_rotr64)
 #   define ROTR(a,n)	_rotr64((a),n)
 #  endif
 #  if defined(_M_IX86) && !defined(OPENSSL_NO_ASM) && !defined(OPENSSL_NO_INLINE_ASM)
@@ -361,6 +369,9 @@ static const SHA_LONG64 K512[80] = {
     }
 #   endif
 #   define PULL64(x) __pull64be(&(x))
+#   if _MSC_VER<=1200
+#    pragma inline_depth(0)
+#   endif
 #  endif
 # endif
 #endif
@@ -382,27 +393,74 @@ static const SHA_LONG64 K512[80] = {
 #define Ch(x,y,z)	(((x) & (y)) ^ ((~(x)) & (z)))
 #define Maj(x,y,z)	(((x) & (y)) ^ ((x) & (z)) ^ ((y) & (z)))
 
-#if defined(OPENSSL_IA32_SSE2) && !defined(OPENSSL_NO_ASM) && !defined(I386_ONLY)
-#define	GO_FOR_SSE2(ctx,in,num)		do {		\
-	void	sha512_block_sse2(void *,const void *,size_t);	\
-	if (!(OPENSSL_ia32cap_P & (1<<26))) break;	\
-	sha512_block_sse2(ctx->h,in,num); return;	\
-					} while (0)
+
+#if defined(__i386) || defined(__i386__) || defined(_M_IX86)
+/*
+ * This code should give better results on 32-bit CPU with less than
+ * ~24 registers, both size and performance wise...
+ */
+static void sha512_block_data_order (SHA512_CTX *ctx, const void *in, size_t num)
+	{
+	const SHA_LONG64 *W=in;
+	SHA_LONG64	A,E,T;
+	SHA_LONG64	X[9+80],*F;
+	int i;
+
+			while (num--) {
+
+	F    = X+80;
+	A    = ctx->h[0];	F[1] = ctx->h[1];
+	F[2] = ctx->h[2];	F[3] = ctx->h[3];
+	E    = ctx->h[4];	F[5] = ctx->h[5];
+	F[6] = ctx->h[6];	F[7] = ctx->h[7];
+
+	for (i=0;i<16;i++,F--)
+		{
+#ifdef B_ENDIAN
+		T = W[i];
+#else
+		T = PULL64(W[i]);
 #endif
+		F[0] = A;
+		F[4] = E;
+		F[8] = T;
+		T   += F[7] + Sigma1(E) + Ch(E,F[5],F[6]) + K512[i];
+		E    = F[3] + T;
+		A    = T + Sigma0(A) + Maj(A,F[1],F[2]);
+		}
+
+	for (;i<80;i++,F--)
+		{
+		T    = sigma0(F[8+16-1]);
+		T   += sigma1(F[8+16-14]);
+		T   += F[8+16] + F[8+16-9];
+
+		F[0] = A;
+		F[4] = E;
+		F[8] = T;
+		T   += F[7] + Sigma1(E) + Ch(E,F[5],F[6]) + K512[i];
+		E    = F[3] + T;
+		A    = T + Sigma0(A) + Maj(A,F[1],F[2]);
+		}
+
+	ctx->h[0] += A;		ctx->h[1] += F[1];
+	ctx->h[2] += F[2];	ctx->h[3] += F[3];
+	ctx->h[4] += E;		ctx->h[5] += F[5];
+	ctx->h[6] += F[6];	ctx->h[7] += F[7];
+
+			W+=SHA_LBLOCK;
+			}
+	}
 
-#ifdef OPENSSL_SMALL_FOOTPRINT
+#elif defined(OPENSSL_SMALL_FOOTPRINT)
 
-static void sha512_block (SHA512_CTX *ctx, const void *in, size_t num)
+static void sha512_block_data_order (SHA512_CTX *ctx, const void *in, size_t num)
 	{
 	const SHA_LONG64 *W=in;
 	SHA_LONG64	a,b,c,d,e,f,g,h,s0,s1,T1,T2;
 	SHA_LONG64	X[16];
 	int i;
 
-#ifdef GO_FOR_SSE2
-	GO_FOR_SSE2(ctx,in,num);
-#endif
-
 			while (num--) {
 
 	a = ctx->h[0];	b = ctx->h[1];	c = ctx->h[2];	d = ctx->h[3];
@@ -447,23 +505,19 @@ static void sha512_block (SHA512_CTX *ctx, const void *in, size_t num)
 	h = Sigma0(a) + Maj(a,b,c);			\
 	d += T1;	h += T1;		} while (0)
 
-#define	ROUND_16_80(i,a,b,c,d,e,f,g,h,X)	do {	\
-	s0 = X[(i+1)&0x0f];	s0 = sigma0(s0);	\
-	s1 = X[(i+14)&0x0f];	s1 = sigma1(s1);	\
-	T1 = X[(i)&0x0f] += s0 + s1 + X[(i+9)&0x0f];	\
-	ROUND_00_15(i,a,b,c,d,e,f,g,h);		} while (0)
+#define	ROUND_16_80(i,j,a,b,c,d,e,f,g,h,X)	do {	\
+	s0 = X[(j+1)&0x0f];	s0 = sigma0(s0);	\
+	s1 = X[(j+14)&0x0f];	s1 = sigma1(s1);	\
+	T1 = X[(j)&0x0f] += s0 + s1 + X[(j+9)&0x0f];	\
+	ROUND_00_15(i+j,a,b,c,d,e,f,g,h);		} while (0)
 
-static void sha512_block (SHA512_CTX *ctx, const void *in, size_t num)
+static void sha512_block_data_order (SHA512_CTX *ctx, const void *in, size_t num)
 	{
 	const SHA_LONG64 *W=in;
 	SHA_LONG64	a,b,c,d,e,f,g,h,s0,s1,T1;
 	SHA_LONG64	X[16];
 	int i;
 
-#ifdef GO_FOR_SSE2
-	GO_FOR_SSE2(ctx,in,num);
-#endif
-
 			while (num--) {
 
 	a = ctx->h[0];	b = ctx->h[1];	c = ctx->h[2];	d = ctx->h[3];
@@ -505,16 +559,24 @@ static void sha512_block (SHA512_CTX *ctx, const void *in, size_t num)
 	T1 = X[15] = PULL64(W[15]);	ROUND_00_15(15,b,c,d,e,f,g,h,a);
 #endif
 
-	for (i=16;i<80;i+=8)
+	for (i=16;i<80;i+=16)
 		{
-		ROUND_16_80(i+0,a,b,c,d,e,f,g,h,X);
-		ROUND_16_80(i+1,h,a,b,c,d,e,f,g,X);
-		ROUND_16_80(i+2,g,h,a,b,c,d,e,f,X);
-		ROUND_16_80(i+3,f,g,h,a,b,c,d,e,X);
-		ROUND_16_80(i+4,e,f,g,h,a,b,c,d,X);
-		ROUND_16_80(i+5,d,e,f,g,h,a,b,c,X);
-		ROUND_16_80(i+6,c,d,e,f,g,h,a,b,X);
-		ROUND_16_80(i+7,b,c,d,e,f,g,h,a,X);
+		ROUND_16_80(i, 0,a,b,c,d,e,f,g,h,X);
+		ROUND_16_80(i, 1,h,a,b,c,d,e,f,g,X);
+		ROUND_16_80(i, 2,g,h,a,b,c,d,e,f,X);
+		ROUND_16_80(i, 3,f,g,h,a,b,c,d,e,X);
+		ROUND_16_80(i, 4,e,f,g,h,a,b,c,d,X);
+		ROUND_16_80(i, 5,d,e,f,g,h,a,b,c,X);
+		ROUND_16_80(i, 6,c,d,e,f,g,h,a,b,X);
+		ROUND_16_80(i, 7,b,c,d,e,f,g,h,a,X);
+		ROUND_16_80(i, 8,a,b,c,d,e,f,g,h,X);
+		ROUND_16_80(i, 9,h,a,b,c,d,e,f,g,X);
+		ROUND_16_80(i,10,g,h,a,b,c,d,e,f,X);
+		ROUND_16_80(i,11,f,g,h,a,b,c,d,e,X);
+		ROUND_16_80(i,12,e,f,g,h,a,b,c,d,X);
+		ROUND_16_80(i,13,d,e,f,g,h,a,b,c,X);
+		ROUND_16_80(i,14,c,d,e,f,g,h,a,b,X);
+		ROUND_16_80(i,15,b,c,d,e,f,g,h,a,X);
 		}
 
 	ctx->h[0] += a;	ctx->h[1] += b;	ctx->h[2] += c;	ctx->h[3] += d;
@@ -528,4 +590,10 @@ static void sha512_block (SHA512_CTX *ctx, const void *in, size_t num)
 
 #endif /* SHA512_ASM */
 
-#endif /* OPENSSL_NO_SHA512 */
+#else /* !OPENSSL_NO_SHA512 */
+
+#if defined(PEDANTIC) || defined(__DECC) || defined(OPENSSL_SYS_MACOSX)
+static void *dummy=&dummy;
+#endif
+
+#endif /* !OPENSSL_NO_SHA512 */